Utility Models
Beyond the conversational and coding tiers, AVELIN-API exposes specialized models for embeddings, transcription, text-to-speech, and image generation. All are available through the same base URL and API key.
Embeddings — bge-m3
High-quality multilingual text embeddings for search, retrieval, and RAG.
curl https://api.avelin.ai/v1/embeddings \ -H "Authorization: Bearer sk-avelin-xxxxxxxxxxxxxxxx" \ -H "Content-Type: application/json" \ -d '{ "model": "bge-m3", "input": "text to embed" }'
Best for: semantic search, document retrieval, clustering, and powering Y-RAY knowledge workflows.
Transcription — whisper-large-v3, whisper-large-v3-turbo, avelin-stt
Speech-to-text transcription.
Endpoint: POST /v1/audio/transcriptions
curl https://api.avelin.ai/v1/audio/transcriptions \ -H "Authorization: Bearer sk-avelin-xxxxxxxxxxxxxxxx" \ -F model="whisper-large-v3-turbo" \ -F file="@meeting.mp3"
Best for: meeting notes, call transcription, voice interfaces.
Text-to-Speech — tts-1, tts-1-hd
Voice synthesis with OpenAI-compatible model and voice names — existing OpenAI SDK clients work unchanged.
Endpoint: POST /v1/audio/speech. Voices follow OpenAI naming (alloy, onyx, ...);
output formats: mp3, opus, pcm.
curl https://api.avelin.ai/v1/audio/speech \ -H "Authorization: Bearer sk-avelin-xxxxxxxxxxxxxxxx" \ -H "Content-Type: application/json" \ -d '{ "model": "tts-1", "voice": "alloy", "input": "Welcome to AVELIN." }' \ --output speech.mp3
Best for: voice interfaces, narration, and audio briefings.
Image Generation — avelin-imagegen, avelin-imagegen-pro
Text-to-image generation.
Endpoint: POST /v1/images/generations
curl https://api.avelin.ai/v1/images/generations \ -H "Authorization: Bearer sk-avelin-xxxxxxxxxxxxxxxx" \ -H "Content-Type: application/json" \ -d '{ "model": "avelin-imagegen", "prompt": "a minimalist product logo" }'
Best for: marketing assets, mockups, and creative concepting.